OBJECTIVE: To monitor the development of 193 children with a psychotic parent genetically at high risk for adult psychosis (experimental group) through a period of three years and to compare and contrast their development with 98 children of physically ill (mainly tuberculous) parents (comparison group) and with 138 children of parents who have experienced no significant mental or physical illness (control group), making a total of 429 children. All of these, currently ranging in age from 7 to 31 years, together with every member of the intact nuclear family group, have already been intensively and extensively studied along a number of different parameters (demographic, physical, psychophysiological, "anthropological", psychiatric, psychological, experimental, and social), and based on these, prognoses have been made as to immediate and remote expectations of psychiatric disorder. The oldest children in all three groups (19 to 31 years and 186 in all) will be passing through a critical period for the genesis of certain types of psychosis, the incidence of which is less than 1 percent for the general population. Three index members have already sustained fully documented "breakdowns" with hospitalizations, whereas the incidence in the comparison and control groups has so far been nil. These prospective investigations of psychosis in the making should throw light on the antecedent variables involved in prediction and help in the early recognition of vulnerable subjects. About one-third of the experimental group has undergone some form of preventive intervention, the efficacy of which is also under consideration.